Mathematical morphology filtering for diffusion tensor MRI

Purpose: Modern imaging such as diffusion tensor magnetic resonance imaging (DT-MRI) is consists of matrix valued data. The processing for DT-MRI tractography is a method to determine the architecture of axonal fibers in the central nervous system by computing the direction of the principal eigenvectors that obtained from tensor matrix. This tensor matrix is usually sensitive to noise. Hence adequate image processing methods for regularization are in demand. Our goal of this paper is extend the morphological operation to the matrix-valued data and applies it to noise reduction. Methods: We proposed three Mathematical Morphology methods for regularization and compared to matrix medians which are the Simple Median Method. Results The results of the Mathematical Morphology Method are faster and the performance is similar to those of the Simple Median. Conclusions: Mathematical Morphology methods are effective algorithm for the tensor regularization.